[GHC] #15052: DeriveAnyClass instances may skip TypeError constraints

[GHC]

#15052: DeriveAnyClass instances may skip TypeError constraints
-------------------------------------+-------------------------------------
        Reporter:  jcpetruzza        |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:  8.6.1
       Component:  Compiler          |              Version:  8.2.2
      Resolution:                    |             Keywords:
                                     |  CustomTypeErrors
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by RyanGlScott):

 You've only presented me with the options that all three should be
 accepted, or all three should be rejected. I don't agree with either of
 those options. I think that, given the current established conventions for
 how `TypeError` works, the first option should be accepted, and the latter
 two options should be rejected. The first option is working in a
 fundamentally different way than the latter two, so it's foolhardy to try
 and equate them.

 I think there might be a different semantics for `TypeError` which would
 reject the first option. But I'd like to put an emphasis on "different".

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15052#comment:12>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler